training_checkpoints/checkpoint-1286/trainer_state.json

{
  "best_metric": 0.5473679856051097,
  "best_model_checkpoint": "tiny-mistral/checkpoint-1286",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1286,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.78,
      "grad_norm": 13.069233894348145,
      "learning_rate": 4.7413167444271646e-05,
      "loss": 1.4479,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6498838109992254,
      "eval_f1_macro": 0.45467871431664447,
      "eval_f1_micro": 0.6498838109992254,
      "eval_f1_weighted": 0.6214028764904677,
      "eval_loss": 1.118202805519104,
      "eval_macro_fpr": 0.039027492204575735,
      "eval_macro_sensitivity": 0.4743680961365506,
      "eval_macro_specificity": 0.9731278674075015,
      "eval_precision": 0.6258370043663808,
      "eval_precision_macro": 0.4712458606651466,
      "eval_recall": 0.6498838109992254,
      "eval_recall_macro": 0.4743680961365506,
      "eval_runtime": 52.8829,
      "eval_samples_per_second": 24.412,
      "eval_steps_per_second": 3.063,
      "eval_weighted_fpr": 0.03705525495982948,
      "eval_weighted_sensitivity": 0.6498838109992254,
      "eval_weighted_specificity": 0.9470342001132973,
      "step": 643
    },
    {
      "epoch": 1.56,
      "grad_norm": 42.20966720581055,
      "learning_rate": 4.482115085536548e-05,
      "loss": 0.8133,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6986831913245546,
      "eval_f1_macro": 0.5473679856051097,
      "eval_f1_micro": 0.6986831913245546,
      "eval_f1_weighted": 0.6969580552726153,
      "eval_loss": 1.0854355096817017,
      "eval_macro_fpr": 0.030488809177152586,
      "eval_macro_sensitivity": 0.552765390116517,
      "eval_macro_specificity": 0.9773020581125585,
      "eval_precision": 0.7196764053623382,
      "eval_precision_macro": 0.5876945028566362,
      "eval_recall": 0.6986831913245546,
      "eval_recall_macro": 0.552765390116517,
      "eval_runtime": 93.8123,
      "eval_samples_per_second": 13.762,
      "eval_steps_per_second": 1.727,
      "eval_weighted_fpr": 0.029883997848966736,
      "eval_weighted_sensitivity": 0.6986831913245546,
      "eval_weighted_specificity": 0.9608476803638234,
      "step": 1286
    }
  ],
  "logging_steps": 500,
  "max_steps": 9645,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 500,
  "total_flos": 5727044348411904.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}